Re:Princeton’s Ad-Blocker May Put an End to the Ad-Blocking ...
by RangerScience on Friday April 14, 2017 @04:06PM
(#14116864)
Check out "General Adversarial Networks". It's basically training a two-part AI via an arms race; one part is trying to fake out the other, the other is trying to detect the fake.
[ Parent
| on HN ]
Re:Princeton’s Ad-Blocker May Put an End to the Ad-Blocking ...
by kalid on Friday April 14, 2017 @04:41PM
(#14117162)
Small fix, I believe it's "Generative" Adversarial Networks.
Re:Princeton’s Ad-Blocker May Put an End to the Ad-Blocking ...
by RangerScience on Friday April 14, 2017 @07:59PM
(#14118440)
You are correct sir.
Re:Princeton’s Ad-Blocker May Put an End to the Ad-Blocking ...
by visarga on Friday April 14, 2017 @11:49PM
(#14119276)
> Check out "General Adversarial Networks"
Another way to describe GANs is that the loss function, which used to be hand-made, now is a whole separate neural net and can be learned bottom-up instead of dictated top-down.
There was a time when features were hand-made. Now we have deep learning and finding good features is automatic. GANs automate another part of the process.
Re:Princeton’s Ad-Blocker May Put an End to the Ad-Blocking ...
by alanbernstein on Friday April 14, 2017 @04:43PM
(#14117186)
So you start a company to produce two networks with GAN techniques, then sell them as separate products, to the opposing sides...